Intestinal digesta is commonly used for studying responses of microbiota to dietary shifts, yet evidence is accumulating that it represents an incomplete view of the intestinal microbiota. In a 16-week seawater feeding trial, Atlantic salmon (Salmo salar) were fed either a commercially-relevant reference diet or an insect meal diet containing ~15% black soldier fly (Hermetia illucens) larvae meal. The digesta- and mucosa-associated distal intestinal microbiota were profiled by 16S rRNA gene sequencing. Regardless of diet, we observed substantial differences between digesta- and mucosa-associated intestinal microbiota. Microbial richness and diversity were much higher in the digesta than the mucosa. The insect meal diet altered the distal intestinal microbiota resulting in higher microbial richness and diversity. The diet effect, however, depended on the sample origin. Digesta-associated intestinal microbiota showed more pronounced changes than the mucosa-associated microbiota. Lastly, multivariate association analyses identified two mucosa-enriched taxa, Brevinema andersonii and unclassified Spirochaetaceae, associated with the expression of genes related to immune responses and barrier function in the distal intestine, respectively. Overall, our data clearly indicate that responses in digesta- and mucosa-associated microbiota to dietary inclusion of insect meal differ, with the latter being more resilient to dietary changes.

Here's an overview of the file organization in this project.

root
├── code                              # all the scripts used for the analysis
│   ├── functions                     # functions for automating tasks
│   ├── utilities                     # utility scripts for miscellaneous tasks
│   ├── 00_setup.bash                 # download raw and reference data for the analysis
│   ├── 01_dada2.Rmd                  # sequence denoising by the dada2 pipeline
│   ├── 02_qiime2_part1.bash          # taxonomic assignment in qiime2
│   ├── 03_preprocessing.Rmd          # feature table filtering    
│   ├── 04_qiime2_part2.bash          # phylogeny and core-metrics-results
│   ├── 05_qiime2R.Rmd                # export qiime2 artifacts into R
│   ├── 06_taxonomy.Rmd               # taxonomic analysis
│   ├── 07_alpha-diversity.Rmd        # alpha-diversity visualization and statistical analysis
│   ├── 08_beta-diversity.Rmd         # beta-diversity visualization and statistical analysis
│   ├── 09_metadata_association.Rmd   # association testing between microbial clades and sample metadata
│   └── README.md
├── data               # all the data, including raw, reference and intermediate data
│   ├── metadata.tsv   # sample metadata
│   ├── raw            # raw data
│   ├── reference      # reference data
│   ├── qPCR           # qPCR assay reports, plat-calibration and Cq values
│   ├── dada2          # outputs from dada2 including the representative sequences and feature table
│   ├── qiime2         # outputs from qiime2
│   ├── preprocessing  # plots for the identification of contaminants; filtered feature table   
│   ├── qiime2R        # RData containing outputs from qiime2
│   ├── permanova      # input data and results of the PERMANOVA
│   └── maaslin2       # default outputs from the maaslin2 program
├── image   # pictures/photos relevant to the analysis
├── result  # final results published with the paper
│   ├── figures    
│   ├── tables     
│   └── README.md 
├── LICENSE.md  
└── README.md

Computationally lightweight RMarkdown files ([03, 05-09]_*.Rmd) can be directly run online by clicking the badge located at the top of this README file. After clicking the badge, this repository will be turned into an RStudio instance that has all the dependencies installed. The instance has limited computational resources (1 cpu; 1~2GB RAM) and is not intended for running tasks requiring intensive computation, i.e., 01_dada2.Rmd.

To reproduce the figures and tables published with the paper, run the following RMarkdown files:

• 03_preprocessing.Rmd
  • Figure S1
  • Table S1
• 06_taxonomy.Rmd
  • Figure 1-2, Figure S2
  • Table S2
• 07_alpha-diversity.Rmd
  • Figure 3
• 08_beta-diversity.Rmd
  • Figure 4
  • Table 2-3
• 09_metadata_association.Rmd
  • Figure 5, Figure S3-8

• Miniconda3 should be located in your HOME directory.
• grabseqs (0.7.0) should be installed via the Miniconda3.
• QIIME2 (2020.2) should be installed within a Miniconda3 environment named as qiime2-2020.2.
  • QIIME2 library: DEICODE (0.2.3) should be installed within the conda environment of qiime2 (qiime2-2020.2).
• Pandoc (1.12.4.2) should be located in your PATH.
• R (3.6.3) should be located in your PATH.
• R packages (packageName_version[source]):
  • ape_5.3 [CRAN]
  • biomformat_1.14.0 [Bioconductor 3.10]
  • circlize_0.4.8 [CRAN]
  • ComplexHeatmap_2.2.0 [Bioconductor 3.10]
  • cowplot_1.0.0 [CRAN]
  • dada2_1.14.1 [Bioconductor 3.10]
  • DT_0.11 [CRAN]
  • EMAtools_0.1.3 [CRAN]
  • emmeans_1.4.4 [CRAN]
  • factoextra_1.0.6 [CRAN]
  • ggResidpanel_0.3.0 [CRAN]
  • ggsignif_0.6.0 [CRAN]
  • ggstatsplot_0.3.1 [CRAN]
  • gridExtra_2.3 [CRAN]
  • gt_0.1.0 [github::rstudio/gt@f793b33]
  • here_0.1 [CRAN]
  • knitr_1.27 [CRAN]
  • lmerTest_3.1-1 [CRAN]
  • lsr_0.5 [CRAN]
  • Maaslin2_1.0.0 [github::biobakery/Maaslin2@RELEASE_3_10]
  • MicrobeR_0.3.1 [github::jbisanz/MicrobeR@7207507]
  • microbiome_1.8.0 [Bioconductor 3.10]
  • PerformanceAnalytics_1.5.3 [CRAN]
  • philr_1.12.0 [Bioconductor 3.10]
  • phyloseq_1.30.0 [Bioconductor 3.10]
  • picante_1.8 [CRAN]
  • plotly_4.9.1 [CRAN]
  • qiime2R_0.99.13 [github::jbisanz/qiime2R@cd07f40]
  • RColorBrewer_1.1-2 [CRAN]
  • rlang_0.4.4 [CRAN]
  • rmarkdown_2.1 [CRAN]
  • scales_1.1.0 [CRAN]
  • tidyverse_1.3.0 [CRAN]
  • vegan_2.5-6 [CRAN]
  • venn_1.9 [CRAN]

All the codes should be run from the project's root directory.

1.Download or clone this github repository to your project's root directory.

# clone the github repository
git clone https://github.com/yanxianl/Li_AqFl2-Microbiota_ASM_2020.git

# delete the following folders which would otherwise cause problems when running `04_qiime2_part2.bash`
rm -rf data/qiime2/core-metrics-results/ data/qiime2/robust-Aitchison-pca/

2.Download the raw sequence data and reference database/phylogenetic tree for the analysis.

bash code/00_setup.bash

3.Sequence denoising by dada2.

Rscript -e "rmarkdown::render('code/01_dada2.Rmd')"

4.Taxonomic assignment in qiime2.

bash code/02_qiime2_part1.bash

5.Filter the feature table to remove: 1).chloroplast/mitochondria sequences and those without a phylum-level taxonomic annotation; 2).low-prevalence features that only present in one sample; 3).contaminating features.

Rscript -e "rmarkdown::render('code/03_preprocessing.Rmd')"

6.Phylogeny and core-metrics-results in qiime2.

bash code/04_qiime2_part2.bash

7.Export qiime2 outputs into R.

Rscript -e "rmarkdown::render('code/05_qiime2R.Rmd')"

8.Taxonomic analysis.

Rscript -e "rmarkdown::render('code/06_taxonomy.Rmd')"

9.Alpha-diversity visualization and statistical analysis.

Rscript -e "rmarkdown::render('code/07_alpha-diversity.Rmd')"

10.Beta-diversity visualization and statistical analysis.

Rscript -e "rmarkdown::render('code/08_beta-diversity.Rmd')"

11.Association testing between microbial clades and sample metadata.

Rscript -e "rmarkdown::render('code/09_metadata_association.Rmd')"

• Add a driver script to automate all the analysis, e.g., make or snakemake.

The initial file and directory structure of this project is based on the template shared by Dr. Pat Schloss to improve the reproducibility of microbiome data analysis. For trainings and tutorials on reproducible data analysis in microbiome research, check the Riffomonas project.

The R package holepunch was used to make the RMarkdown files binder-ready.